- About CSAIL
- News + Events
- Alumni & Friends
Graphics and Vision
Multiobjective Control with Frictional Contacts
Standing is a fundamental skill mastered by humans and animals alike. Although easy for adults, it requires careful and deliberate manipulation of contact forces. The variation in contact confguration (e.g., standing on one foot, on uneven ground, or while holding on for support) presents a diffcult challenge for interactive simulation of humans and animals, especially while performing tasks in the presence of external disturbances. We describe an analytic approach for control of standing in three-dimensional simulations based upon local optimization. At any point in time, the control system solves a quadratic program to compute actuation by maximizing the performance of multiple motion objectives subject to constraints imposed by actuation limits and contact configuration.
Guided Time Warping for Motion Editing
Time warping allows users to modify timing without affecting poses. It has many applications in animation systems for motion editing, such as refining motions to meet new timing constraints or modifying the acting of animated characters. However, time warping typically requires many manual adjustments to achieve the desired results. We present a technique which simplifies this process by allowing time warps to be guided by a provided reference motion. Given few timing constraints, it computes a warp that both satisfies these constraints and maximizes local timing similarities to the reference. The algorithm is fast enough to incorporate into standard animation workflows.
A Topological Approach to Hierarchical Segmentation Using Mean Shift
Mean shift is a popular method to segment images and videos. Pixels are represented by feature points, and the segmentation is driven by the point density in feature space. In this paper, we introduce the use of Morse theory to interpret mean shift as a topological decomposition of the feature space into density modes. This allows us to build on the watershed technique and design a new algorithm to compute mean-shift segmentations of images and videos. In addition, we introduce the use of topological persistence to create a segmentation hierarchy. We validated our method by clustering images using color cues. In this context, our technique runs faster than previous work, especially on videos and large images.
Two-scale Tone Management for Photographic Look
We introduce a new approach to tone management for photographs. Whereas traditional tone-mapping operators target a neutral and faithful rendition of the input image, we explore pictorial looks by controlling visual qualities such as the tonal balance and the amount of detail. Our method is based on a two-scale non-linear decomposition of an image. We modify the different layers based on their histograms and introduce a technique that controls the spatial variation of detail. We introduce a Poisson correction that prevents potential gradient reversal and preserves detail. In addition to directly controlling the parameters, the user can transfer the look of a model photograph to the picture being edited.
Inverse Kinematics for Reduced Deformable Models
Articulated shapes are aptly described by reduced deformable models that express required shape deformations using a compact set of control parameters. Although sufficient to describe most shape deformations, these control parameters can be ill-suited for animation tasks, particularly when reduced deformable models are inferred automatically from example shapes.
Interactive Animation of Dynamic Manipulation
Lifelike animation of object manipulation requires dynamic interaction between animated characters, objects, and their environment. These interactions can be animated automatically with physically based simulations but proper controls are needed to animate characters that move realistically and that accomplish tasks in spite of unexpected disturbances. This paper describes an effcient control algorithm that generates realistic animations by incorporating motion data into task execution. The end result is a versatile system for interactive animation of dynamic manipulation tasks such as lifting, catching, and throwing.
Face Transfer with Multilinear Models
Face Transfer is a method for mapping videorecorded performances of one individual to facial animations of another. It extracts visemes (speech-related mouth articulations), expressions, and three-dimensional (3D) pose from monocular video or film footage. These parameters are then used to generate and drive a detailed 3D textured face mesh for a target identity, which can be seamlessly rendered back into target footage. The underlying face model automatically adjusts for how the target performs facial expressions and visemes. The performance data can be easily edited to change the visemes, expressions, pose, or even the identity of the target—the attributes are separably controllable.
Continuous Capture of Skin Deformation
We describe a method for the acquisition of deformable human geometry from silhouettes. Our technique uses a commercial tracking system to determine the motion of the skeleton, then estimates geometry for each bone using constraints provided by the silhouettes from one or more cameras. These silhouettes do not give a complete characterization of the geometry for a particular point in time, but when the subject moves, many observations of the same local geometries allow the construction of a complete model. Our reconstruction algorithm provides a simple mechanism for solving the problems of view aggregation, occlusion handling, hole filling, noise removal, and deformation modeling. The resulting model is parameterized to synthesize geometry for new poses of the skeleton.
Style Translation for Human Motion
Style translation is the process of transforming an input motion into a new style while preserving its original content. This problem is motivated by the needs of interactive applications, which require rapid processing of captured performances. Our solution learns to translate by analyzing differences between performances of the same content in input and output styles. It relies on a novel correspondence algorithm to align motions, and a linear time-invariant model to represent stylistic differences. Once the model is estimated with system identification, our system is capable of translating streaming input with simple linear operations at each frame.